Optimal selection of controlled variables

This paper considers the selection of controlled variables for the unconstrained degrees of freedom, such that near-optimal operation is achieved with constant setpoints ("self-optimizing control"). From a second-order Taylor approximation around the optimal point, we derive an exact local method. This may be used to find the optimal linear combination of measurements to use as controlled variables. We also derive the approximate singular value rule, which is very useful for quick screening and elimination of poor candidate variables.